final commit

app.py

@@ -17,10 +17,10 @@ def calc_voxsim(inp_path, ref_path):
     ref_wav = torch.FloatTensor(ref_wav)
 
     with torch.no_grad():
-        input_emb_1 = F.normalize(model.foward(inp_wavs), p=2, dim=1)
-        input_emb_2 = F.normalize(model.foward(inp_wav), p=2, dim=1)
-        ref_emb_1 = F.normalize(model.foward(ref_wavs), p=2, dim=1)
-        ref_emb_2 = F.normalize(model.foward(ref_wav), p=2, dim=1)
+        input_emb_1 = F.normalize(model.forward(inp_wavs), p=2, dim=1)
+        input_emb_2 = F.normalize(model.forward(inp_wav), p=2, dim=1)
+        ref_emb_1 = F.normalize(model.forward(ref_wavs), p=2, dim=1)
+        ref_emb_2 = F.normalize(model.forward(ref_wav), p=2, dim=1)
 
         score_1 = torch.mean(torch.matmul(input_emb_1, ref_emb_1.T))
         score_2 = torch.mean(torch.matmul(input_emb_2, ref_emb_2.T))

requirements.txt

@@ -4,4 +4,5 @@ torch
 torchaudio
 tqdm
 s3prl
-huggingface_hub
+huggingface_hub
+gradio